Network (Function) Virtualization and Fog Service Support in 5G

Cloud and Edge/Fog Computing (EC/FC), Network Function Virtualization (NFV), and Software Defined Networking (SDN) have stood out as important technologies that can be used to create new opportunities that will meet the growing computational and networking requirements for upcoming Fifth-Generation/5G (and beyond - B5G) Mobile Wireless Networks.

These technologies are especially important, as 5G networks will need to provide connectivity for a wide range of services with heterogeneous requirements, over a shared underlying infrastructure. These services can be designated into 3 main categories, namely: enhanced Mobile Broadband (eMBB), massive Machine Type Communications (mMTC) and ultra-Reliable Low-Latency Communications (uRLLC). Supporting such services over a common infrastructure is made complicated by the limited capacity of edge/fog devices and the dynamic nature of the edge (i.e. varying channel conditions, and traffic dynamics). Therefore it is important to be able to dynamically adapt the allocation of resources to the services running in such environments to ensure that their quality of service requirements (such as latency, and throughput) are met.

One way to do this, is to apply AI-based approaches in the resource allocation and orchestration process, to enable better performance of such services and network functions, giving the resource constraints and also network uncertainty/stochasticity.

Research in this area considers how to make use of intelligent and heuristic methods (among others) to optimize and scale the allocation of various network resources, with the goal of improving the performance of services in virtualization-enabled 5G networks.

Related Literature:

[1] Optimal Resource Allocation in C-RAN through DSP Computational Load Forecasting

[2] Demystifying the Performance Interference of Co-located Virtual Network Functions

[3] vrAIn: A Deep Learning Approach Tailoring Computing and Radio Resources in Virtualized RANs

[4] Management and Orchestration of Virtual Network Functions via Deep Reinforcement Learning

[5] The Path Towards Resource Elasticity for 5G Network Architecture

[6] Multi-Domain Orchestration for NFV: Challenges and Research Directions